The relationship of facial anatomic

landmarks with midlines of the face and

mouth
Avinash S. Bidra, BDS, MDentSc,a Flavio Uribe, DDS,
MDentSc,b Thomas D. Taylor, DDS, MSD,c John R. Agar, DDS,
MA,d Patchnee Rungruanganunt, DDS, MSD,e and William P.
Neace, PhDf
University of Texas Health Science Center at San Antonio, San
Antonio, Texas; University of Connecticut School of Dental
Medicine, Farmington, Conn; University of Hartford, Hartford,
Conn
Statement of problem. The importance of the midline is well known to dentists. Currently, there are no verifiable
guidelines that direct the choice of specific anatomic landmarks to determine the midline of the face or midline of the
mouth.

Purpose. The purpose of this study was to determine the hierarchy of facial anatomic landmarks closest to the midline
of the face as well as midline of the mouth.

Material and methods. Three commonly used anatomic landmarks, nasion, tip of the nose, and tip of the philtrum,
were marked clinically on 249 subjects (age range: 21-45 years). Frontal full-face digital images of the subjects in smile
were then made under standardized conditions. A total of 107 subjects met the inclusion criteria. Upon applying ex-
clusion criteria, images of 87 subjects were used for midline analysis using a novel concept called the Esthetic Frame.
Deviations from the midlines of the face and mouth were measured for the 3 clinical landmarks; the existing dental
midline was considered as the fourth landmark. The entire process of midline analysis was done by a single observer
and repeated twice. Reliability analysis and 1-sample t tests were conducted at alpha values of .001 and .05, respec-
tively.

Results. The results indicated that each of the 4 landmarks deviated uniquely and significantly (P<.001) from the mid-
lines of the face as well as the mouth.

Conclusions. Within the limitations of the study, the hierarchy of anatomic landmarks closest to the midline of the
face in smile was as follows: the midline of the oral commissures, natural dental midline, tip of philtrum, nasion, and
tip of the nose. The hierarchy of anatomic landmarks closest to the midline of the oral commissures was: natural
dental midline, tip of philtrum, tip of the nose, and nasion. These relationships were the same for both genders and all
ethnicities classified. (J Prosthet Dent 2009;102:94-103)

Presented as a table clinic at the American Academy of Fixed Prosthodontics meeting, February 2009.
Supported by the Tylman Grant award, American Academy of Fixed Prosthodontics.
Third place winner of the Academy of Prosthodontics Foundation award.

a
Assistant Professor, Department of Prosthodontics, University of Texas Health Science Center at San Antonio; Former Resident,
Graduate Prosthodontics, University of Connecticut School of Dental Medicine.
b
Assistant Professor and Program Director, Division of Orthodontics, University of Connecticut School of Dental Medicine.
c
Professor and Chair, Department of Reconstructive Sciences, University of Connecticut School of Dental Medicine.
d
Professor and Program Director, Division of Prosthodontics, University of Connecticut School of Dental Medicine.
e
Assistant Professor, Department of Reconstructive Sciences, University of Connecticut School of Dental Medicine.
f
Assistant Professor, Department of Psychology, University of Hartford.

The Journal of Prosthetic Dentistry Bidra et al

August 2009
Clinical Implications
The midline of the oral commissures, natural dental midline, and tip
of the philtrum, in this order, can be chosen as preferential landmarks
in the determination of the midline of the face and mouth in smile.
Symmetry, normalcy, sexual di-
morphism, and youthfulness have
been considered the classical ele-
ments of facial beauty.1,2 By definition,
symmetry is the “correspondence in
size, shape, and relative position of
parts on opposite sides of a dividing
line or median plane or about a cen-
ter or axis.”3 This dividing line, which
is used to attain symmetry, is known
as the midline. It is the fundamental
reference for all esthetic deviations.
Therefore, knowledge of the midline
will invariably result in a better under-
standing of facial and dental esthet-
ics. Historically, a diverse number of
facial anatomic landmarks located on
the middle third of the face, such as
the bisector of the pupils, nasion, tip
of the nose, tip of the philtrum, and
chin, have been used to determine the
facial and dental midlines.4,5 Some
advocate the use of intraoral land-
marks, such as the incisive papilla, for
determination of the maxillary dental
midline.6,7 It has been argued in the
literature whether the dental midline
should be made coincident with the
midline of the face or the midline of
the oral commissures.8 Some believe
that making the dental midline co-
incident with the midline of the oral
commissures is adequate, as patients
tend to relate their dental midline to
proximal structures rather than ana-
tomic structures which are farther
from the mouth.9,10 However, the
literature is not clear regarding veri-
fiable guidelines for the determina-
tion of midlines of the face or mouth.
Based upon convention and dogma,
most clinicians choose one specific
anatomic landmark and an imaginary
line passing through it. Others use
dental floss and hold it in front of the
face from glabella to menton. Thus,
the clinician is left with no predict-
Bidra et al
able guidelines, and must determine
the midline based on unverified land-
marks.
In dental esthetics, it is more im-
portant that the maxillary dental mid-
line and the facial midline coincide,
than the mandibular and facial mid-
lines. This is due to the dominant visi-
bility of the maxillary anterior teeth in
smile and function. The coincidence
of facial, maxillary, and mandibular
midlines is desirable, but not man-
dated. The maxillary dental and facial
midlines have an important role in
esthetics and occlusion in many disci-
plines, including removable and fixed
prosthodontics, implant prosthodon-
tics, orthodontics, and facial plastic
surgery. Not being able to match the
dental midline coincident to the mid-
line of the face or mouth is a common
cause of frustration for dentists. This
generally results in elaborative proce-
dures for correction, causing loss of
clinical time.
Most of the literature available
with regard to this topic is restricted
to monographs written by various
authors. Clinical studies have been
limited to the amount of tolerance
of deviated dental midlines from the
facial midline, a span of approxi-
mately 2 to 3 mm.5,11-14 Textbooks
and monographs written by various
authors reveal a division of thought
with respect to whether or not the
dental midline should be placed ex-
actly coincident to the facial mid-
line.15 Lombardi suggests placing the
dental midline in the location on the
face where it appears most “stable.”8
Farkas16 described the facial midline
in anthropometric interest, as a line
defined by 3 anatomic points: na-
sion, subnasale, and the gnathion
or menton. However, this definition
is not clear and does not lend itself
95
to objectivity and repeatability for
research purposes. The Glossary of
Prosthodontic Terms,17 American As-
sociation of Orthodontists (AAO)
Glossary,18 and the Glossary of Terms
for the American Academy of Facial
Plastic and Reconstructive Surgery19
currently do not have any definitions
for facial and dental midlines. To the
authors’ knowledge, there is no litera-
ture describing the relationship of the
midline of the mouth to the midline
of the face.
Miller et al15 was the first to study
the clinical relationship of the dental
midline and the facial median line,
in a study involving a sample of 500
human subjects. No objective meth-
ods were used in this study. The au-
thors demonstrated that 70.4% of the
sample showed a coincidence of the
dental midline and facial median line.
Owens et al20 conducted a multicenter
study involving 5 centers around the
world to compare facial appearance
across 6 racial groups. Two hundred
fifty-three subjects from the 5 cen-
ters were chosen. Coincidence of the
dental midline with the line bisecting
the interpupillary line was one of the
analyzed variables. The results of the
study showed that 70% of all subjects
had their dental midline coincident
with the line perpendicular to the in-
terpupillary line. Latta7 studied the
relationship between facial midline
and intraoral landmarks in 100 pa-
tients requiring maxillary dentures.
The philtrum was used to represent
the facial midline and its marking was
transferred to a cast manually using
an occlusal rim. The results showed
that the incisive papilla was the clos-
est landmark, followed by the maxil-
lary frenum and the midpalatal su-
ture. All of the previously mentioned
studies have lacked objectivity in the
96
evaluation criteria for facial midlines.
Thus, the gaps in knowledge in
this field are the lack of a repeatable
and verifiable definition for facial and
dental midlines and lack of scientific
information on relationships of spe-
cific anatomic landmarks with the
facial midline. Furthermore, there is
a lack of information about the rela-
tionship of the 2 standard midlines:
midline of the face and the midline of
the mouth. All of these provided the
rationale for this study.
The objectives of the study were
to define: (1) the hierarchy of facial
anatomic landmarks closest to the
midline of the face; (2) the hierarchy
of facial anatomic landmarks closest
to the midline of the oral commis-
sures (mouth); and (3) the relation-
ship between the midline of the oral
commissures and the midline of the
face. The facial anatomic landmarks
analyzed were those traditionally used
in clinical practice such as: nasion, tip
of the nose (pronasale), tip of phil-
trum (labiale superioris), and dental
midline. The null hypothesis was that
there would be no difference between
the chosen facial anatomic landmarks
and the midlines of the face and oral
commissures.
MATERIAL AND METHODS
Institutional Review Board ap-
proval (#07-117-2) was obtained for
convenience sampling of 249 medi-
cal and dental students/employees
with an age range of 21-45 years at
the University of Connecticut Health
Center. Each subject had 3 small
marks placed by a single observer
using a fine-tipped erasable marker,
with a tip approximately 0.5 mm in
diameter (Expo; Sanford Ink Co, Oak
Brook, Ill) on the nasion, tip of the
nose, and tip of the philtrum, to sim-
ulate a clinical situation. Standardiza-
tion was given to application of all
anatomic marks in terms of the stan-
dard anatomic and anthropologic
definitions described below. A digital
camera (Nikon D70s digital camera,
6.1 Megapixel; Nikon USA, Melville,
The Journal of Prosthetic Dentistry
NY) with a 105-mm lens and a point
flash, as well as an additional flash
with a wireless speedlight (Nikon
SB-R200 Wireless Speedlight; Nikon
USA), was used in the 12 o’clock po-
sition. The camera had an aperture
setting of F4.5 and was mounted on
a tripod (Canon Deluxe 200 Tripod;
Canon USA, Lake Success, NY) with a
standardized focus and at a standard-
ized distance of 5 feet (1.5 m) from
the subject. The lighting conditions
remained the same for all the pho-
tographs. This procedure was similar
to the protocol described by Owens
et al.20 Full-face digital images of sub-
jects in smile were made, with the
subject in a seated position. The head
position was guided by the observer
to assist the subjects in assuming their
natural head position, an approach
which has been well documented in
the literature.21-24 The height of the
lens of the camera was adjusted on
the tripod to match the eye level of
the subject when seated upright with
shoulders and head held straight and
facing forward (Fig. 1). As the subject
looked straight ahead at the lens of
the camera on a tripod, the natural
head position was standardized along
both horizontal and vertical axes. As
long as the eyes of a subject were not
naturally located at different levels
in the natural head position, any mi-
nor rotations of the head along the
sagittal axis were nullified when the
intercanthal line was digitally made
parallel to the true horizontal before
analysis. Imaging software (Adobe
1 Schematic illustration showing methodology of obtaining
standard digital images of subject in natural head position.
Volume 102 Issue 2
Photoshop CS2; Adobe Systems, Inc,
San Jose, Calif ) was used to digitally
analyze the photographs.
Upon initial screening of 249 sub-
jects, 142 of them did not meet the
inclusion criteria, which were as fol-
lows: age range of 21-45 years, no
history of any congenital conditions
or trauma affecting facial form and
appearance, no history of orthodon-
tic treatment, no missing maxillary
anterior teeth, no prosthetic maxillary
anterior teeth, no interdental spacing
in the maxillary teeth, ability to under-
stand written informed consent docu-
ments and the verbal explanation.
The inclusion criteria were applied
based on the above data recorded on
a separate sheet for each subject. The
predetermined exclusion criteria were
applied to the images of the 107 in-
cluded subjects. The exclusion criteria
were as follows: images with rotations
of head around the vertical axis, ob-
vious ophthalmic asymmetry, inac-
curate clinical markings, and images
without a good resolution. Upon ap-
plying the exclusion criteria, the total
number of subjects used for midline
analysis was 87 (Tables I and II). All
images were made with the subject’s
head in a natural head position. Prior
to making images, careful attention
was given to ensure that the subjects
did not rotate their heads, especially
along the vertical axis. In addition,
the analysis of subjects with heads ro-
tated along the vertical axis resulted
in the midline being obviously dis-
placed, by a large amount, in a direc-
Bidra et al
August 2009
Table I. Application of exclusion criteria to sample for midline analysis
Exclusion Criteria
Rotation of head along vertical axis
Ophthalmic asymmetry
Inaccurate markings
Unreadable images
Total
Table II. Distribution of sample based on ethnicity and gender
Category
Asians
Blacks
Whites
Other ethnicities
Men
Women
Total
tion opposite to the side of rotation;
hence, these subjects were excluded.
The study excluded subjects who had
obvious ophthalmic asymmetry (eyes
located at different levels) when they
were positioned in their natural head
position. Subjects who were excluded
due to obvious ophthalmic asymme-
try were reanalyzed. The magnitude
of asymmetry was quantified, result-
ing in exclusion of subjects with more
than 1 degree of discrepancy between
the true horizontal line and the in-
tercanthal line. Finally, images which
showed that the clinical markings did
not conform to the standard anatom-
ic definitions at a magnification of at
least 200% were considered inaccu-
rate, and they were excluded from the
study.
Standard definitions for anatomic
landmarks were used for all purposes
of the study. Lateral canthus was de-
fined as the lateral angle formed by
the meeting of the upper and lower
eyelids.3 Exocanthion was defined as
the point at the outer commissure of
the eye fissure.16 Nasion was defined
as the point in the midline of both the
nasal root and nasofrontal suture.3,16
Bidra et al
Number of
Subjects Rejected
8 as perpendicular lines from the exo-
5
5
2 line drawn at the most inferior border
20
Distribution
9 using the “entire” face. It was also as-
9 sumed that the tissues excluded from
50
19
49
38
87
Philtrum was defined as the vertical
groove on the median line of the up-
per lip.3 Commissure was defined as
a point or line of junction between
2 anatomic parts (the lips).3,17 Chei-
lion was defined as the point located
at each labial commissure.16 Tip of
the nose (pronasale) was defined as
the most protruded point of the apex
of the nose.16 These definitions were
used for all clinical markings as well
as to digitally construct an “Esthetic
Frame.”
The Esthetic Frame is novel and
unique to this study. As it is almost
impossible to define the midline of the
face in both static and dynamic move-
ments, the Esthetic Frame comprising
of a rectangular enclosure was used
to define the facial midline objec-
tively. It was defined as an area on the
human face, within which items of es-
thetic interest such as midlines, cants,
and smile parameters are sensitively
perceptible and objectively verifiable.
Its superior border was defined by a
line originating at the exocanthion of
1 eye and meeting the exocanthion of
the other eye. This line helped to ne-
gate the effect of any minor rotations
97
of the head along the sagittal axis.
Subjects with ophthalmic asymmetry
were excluded for analysis using this
frame due to this reason. The 2 lateral
borders of the frame were then drawn
canthion of each eye and were parallel
to each other. The inferior border of
the frame was parallel to the superior
of the lower lip. This completed the 4
sides of the frame (Fig. 2). It was as-
sumed that it was more imperative
to obtain the midline of that portion
of the face included in this Esthetic
Frame, rather than the “true” midline
the Esthetic Frame, such as the chin,
buccal soft tissues, and forehead,
have little to do with the perception
of the facial midline. This is simply
because of the dynamic nature of
the mandible, the irregular hypertro-
phies of the buccinators and masseter
muscles, and the variable size of the
forehead, all of which could poten-
tially serve as confounding variables
in midline perception (Fig. 3).
For this study, the facial midline
was defined as the midline of the es-
thetic frame of the face. The dental
midline was defined as the vertical
line through the tip of the incisal em-
brasure between the 2 maxillary cen-
tral incisors and parallel to the verti-
cal lines of the esthetic frame of the
face. The midline of the oral commis-
sures was defined as a line bisecting
the distance between the cheilions of
the subject in smiling posture. Rela-
tive facial midline value (RFV) and
relative commissural midline value
(RCV) were 2 operational tools used
to quantify the relationships of the
anatomic landmarks to the respec-
tive midlines. The Esthetic Frame was
first constructed on a subject’s im-
age digitally. The facial midline was
established by bisecting the distance
between the 2 lateral borders on the
frame. Three vertical lines were then
drawn along each of the anatomic
points, which had been marked clini-
cally. The fourth line was drawn along
98
2 Computer-generated human face
(FaceGen Modeller; Singular Inver-
sions, Toronto, Ont) to schematically
illustrate concept of Esthetic Frame
used to obtain objectivity and reli-
ability.
the subject’s existing dental midline
as defined above (Fig. 4).
Relative facial midline value (RFV)
was defined as the relative closeness
of an anatomic landmark to the facial
midline. The measured distance from
the lateral border of the frame to the
defined facial midline was considered
a constant called “F.” The measured
distance from the lateral border of
the frame to the nasion was consid-
ered a variable termed “n.” The RFV
was then obtained by dividing n by F.
Similarly, RFVs were obtained for the
other 3 anatomic landmarks: tip of
the nose (t), tip of philtrum (p), and
dental midline (d), by dividing them
by the constant F. Numerical values
for n/F, t/F, p/F, and d/F were thus
obtained (Fig. 5).
Relative commissural midline
value (RCV) was defined as the rela-
tive closeness of an anatomic land-
mark to the midline of the oral com-
missures (center of the mouth). The
measured distance from the midpoint
of the intercommissural line to the
The Journal of Prosthetic Dentistry
3 Tissues excluded from Esthetic
Frame have little to do with percep-
tion of facial midline.
right/left cheilion was considered a
constant termed C. The measured
distances (variables) were: from the
nasion, nx, from the tip of the nose,
tx, from the tip of philtrum, px, and
from the dental midline, dx. The RCV
was then obtained by dividing nx/C,
tx/C, px/C, and dx/C. The measured
distance from the lateral border of the
Esthetic Frame to the midpoint of the
commissures was described as a vari-
able called Cx. Thus, the relationship
between the midline of the commis-
sures and the midline of the face was
obtained by dividing Cx/F (Fig. 6).
The primary reason to use RFV
and RCV was to develop a propor-
tional relationship between an ana-
tomic landmark and the midline in
question. This ensured a standard
common denominator for all ana-
tomic landmarks within the esthetic
frame and negated the need for size
matching the images with the sub-
ject’s face. The assignments for rela-
tivity of landmarks for both midlines
were: RFV1 and RCV1: relativity of
Volume 102 Issue 2
4 Computer-generated human face
seen in Figure 1 was digitally altered
to produce deviated anatomic land-
marks. Figure illustrates methodology
of lines drawn along each deviated
anatomic landmark and determina-
tion of facial midline.
nasion to midline of the face and
commissures; RFV2 and RCV2: rela-
tivity of tip of the nose to midline of
the face and commissures; RFV3 and
RCV3: relativity of tip of the philtrum
to midline of the face and commis-
sures; RFV4 and RCV4: relativity of
dental midline to midline of the face
and commissures; and RFV5: relativ-
ity of the midline of the commissures
with the midline of the face. Thus, in
perfect symmetry, all 5 of the RFVs
and all 4 of the RCVs would be equal
to each other and to the numeral 1.
The right or left lateral border of the
esthetic frame or the commissures
was chosen, based on the direction of
deviation of the anatomic landmark.
Therefore, the shorter distance to the
lateral border of the frame was always
chosen. Thus, an RFV and an RCV
could never be a number greater than
the numeral 1. If a line drawn along
one anatomic landmark coincided
with any of the other landmarks, the
same RFV or RCV value was recorded
for both. If an anatomic landmark
Bidra et al
August 2009 99

5 Method of determination of RFV values for
each anatomic landmark. F: midline of face/mid-
line of esthetic frame; n: distance between nasion
and lateral border of esthetic frame; t: distance
between tip of nose and lateral border of es-
thetic frame; p: distance between tip of philtrum
and lateral border of esthetic frame; d: distance
between dental midline and lateral border of
esthetic frame.
6 Method of determination of RFV5 and RCV
values for each anatomic landmark. F: midline
of face/midline of esthetic frame; C: midline of
oral commissures; Cx: distance between midline
of commissures and lateral border of esthetic
frame; nx: distance between nasion and oral
commissures; tx: distance between tip of nose
and oral commissures; px: distance between tip
of philtrum and oral commissures; dx: distance
between dental midline and oral commissures.

was coincident with the facial or the
commissural midline, then it was as-
signed an RFV or RCV value of 1.
A total of 9 values were recorded
per subject, along with gender and
ethnicity. The entire process of data
analysis was repeated twice to ensure
reliability and validity. A reliability
analysis test was performed between
the first and second set of data us-
ing intraclass correlation coefficients
(ICCs). To determine whether the
selected landmarks significantly dif-
fered from the midline of the face and
mouth, a series of 1-sample t tests
were conducted with an alpha value
of .05. Finally, a Pearson correlation
analysis was performed to determine
whether there was a significant cor-
relation between RCV1 (nasion) and
RCV2 (tip of the nose), as nasion and
tip of the nose showed reversal in hi-
erarchy in relationship to the midline
of the commissures.
Bidra et al
RESULTS
Intraclass correlation coefficients
(ICCs) for reliability analysis of RFV
and RCV measures made 2 times re-
vealed that the reliabilities were all ac-
ceptable, ranging from 0.85 to 0.96,
indicating a high consistency between
measurements made the first and the
second time by the same rater. Items
measured at analysis 1 are paired
with same items measured at analy-
sis 2 (for example, RFV11 = “RFV1”
measured the first time, and RFV12
= “RFV1” measured the second time).
All ICCs were statistically significant
beyond the alpha value of .001 (Table
III).
Two sets of 1-sample t tests were
conducted. One set of 5 t tests was
conducted to test the null hypothesis
that the mean ratios of the 5 speci-
fied anatomic measures did not differ
from 1.00 (whether they all lined up
with the facial midline). The analysis
indicated that the difference between
the mean ratio of each anatomic
landmark and the midline of the face
was statistically significant (P<.001).
The midline of the commissures was
the closest, followed by the dental
midline, tip of philtrum, nasion, and
the tip of the nose (Table IV) (Fig. 7).
A second set of 4 t tests was con-
ducted to test the null hypothesis that
the mean ratios of the 4 specified an-
atomic measures did not differ from
1.00 (whether they all lined up with
the intercommissural midline). In line
with the previous analysis, the results
indicated that the difference between
the mean ratio of each anatomic
landmark and the midline of the com-
missures was statistically significant
(P<.001). The natural dental midline
was the closest, followed by the tip of
philtrum, tip of the nose, and nasion
100 Volume 102 Issue 2

Table III. Reliability analysis table

Item Pair Reliability

RFV11 (nasion) RFV12 0.85

RFV21 (tip of nose) RFV22 0.92
RFV31 (tip of philtrum) RFV32 0.87
RFV41 (dental midline) RFV42 0.85
RFV51 (midline of commissures) RFV52 0.86

RCV11 (nasion) RCV12 0.93

RCV21 (tip of nose) RCV22 0.96
RCV31 (tip of philtrum) RCV32 0.90
RCV41 (dental midline) RCV42 0.94

RFV: Relative facial midline value

RCV: Relative commissural midline value

Table IV. One-sample t test for hierarchy of landmarks for midline of face
Landmark Mean Standard Deviation P

Midline of commissures (RFV5) 0.979 0.017 <.001

Dental midline (RFV4) 0.977 0.018 <.001

Tip of philtrum (RFV3) 0.974 0.021 <.001

Nasion (RFV1) 0.967 0.019 <.001

Tip of nose (RFV2) 0.965 0.025 <.001

RFV: Relative facial midline value

Midline of
face
Midline of
mouth
Dental
midline
Tip of
philtrum

Nasion

Tip of
nose

0.94 0.95 0.96 0.97 0.98 0.99 1.00

7 Hierarchical relationship of anatomic landmarks
with midline of face.

The Journal of Prosthetic Dentistry Bidra et al

August 2009 101

Table V. One-sample t test for hierarchy of landmarks for midline of commissures

Landmark Mean Standard Deviation P

Dental midline (RCV4) 0.9751 0.023 <.001

Tip of philtrum (RCV3) 0.9748 0.024 <.001

Tip of nose (RCV2) 0.9512 0.034 <.001

Nasion (RCV1) 0.9477 0.032 <.001

RCV: Relative commissural midline value

Midline of
mouth
Dental
midline
Tip of
philtrum
Tip of
nose

Nasion

0.94 0.95 0.96 0.97 0.98 0.99 1.00

8 Hierarchical relationship of anatomic landmarks with midline of mouth.

Table VI. Pearson correlation coefficient between RCV1 and RCV2

to demonstrate reversal of hierarchy

RCV1 RCV2

Nasion (RCV1) Pearson correlation 1 0.229

Significance (2-tailed) 0.033
n 87 87

Tip of nose (RCV2) Pearson correlation 0.229 1

Significance (2-tailed) 0.033
n 87 87

RCV: Relative commissural midline value

(Table V) (Fig. 8). These hierarchical
relationships remained the same for
both genders and all ethnicities con-
sidered.
A Pearson correlation analysis was
done to determine any significant
association between the 2 samples,
RCV1 (nasion) and RCV2 (tip of the
nose), as nasion and tip of the nose
showed reversal in hierarchy in rela-
tionship to the midline of the commis-
sures. Table VI shows that RCV1 and
Bidra et al
RCV2 are significantly correlated at a
.05 level (2-tailed),  indicating these
2 values were similar to each other;
hence, the hierarchy is negligible.
DISCUSSION
The results support rejection of
the null hypothesis that there would
be no difference between the chosen
facial anatomic landmarks and the
midlines of the face and oral com-
missures. There is no standard defini-
tion for facial midline in the literature.
Therefore, the authors defined the fa-
cial midline using the Esthetic Frame.
Many authors have shown the validity
of the natural head position and its
long-term reproducibility over a pe-
riod of up to 15 years.21-24 In the pres-
ent study, the natural head position
was guided to the true horizontal by a
single investigator, and care was taken
to ensure that the subjects did not ro-
102
tate their heads along the vertical axis.
However, human error in detection of
this rotation cannot be ruled out. The
smiling image of the subject was cho-
sen for all purposes of analysis, as it is
a standard for esthetic analysis, and
it revealed the dental midline as well.
No subject in this study had a grossly
asymmetric smile or a smile which did
not reveal the maxillary central inci-
sors. All exclusion criteria reported in
this study was stringently applied to
minimize the number of confounding
factors.
This study was designed to be
as clinically applicable as possible;
therefore, the markings for each ana-
tomic landmark were made clinically
and not on the digital image. The
lines on the image were drawn along
these markings. Although meticulous
care and clinical judgment were exer-
cised during the marking procedure,
inherent human errors in marking
the anatomic landmarks clinically
cannot be eliminated. Of the various
clinical landmarks, marking the soft
tissue nasion and the tip of the nose
was most difficult due to the inherent
anatomy of the nose.16 Hence, the re-
sults related to these anatomic land-
marks should be carefully considered,
and future studies are needed to verify
these results. The use of cephalomet-
rics was considered, but not used, as
it would not simulate a clinical situa-
tion for recording the midline.
The midline of the oral commis-
sures was considered as a determined
anatomic landmark while analyzing
the hierarchical order for facial mid-
lines. It ranked the closest to the facial
midline, in comparison to all of the
landmarks analyzed. This may reveal
that nature centers the mouth quite
symmetrically in relationship to the
eyes. The dental midline in this popu-
lation with no history of orthodontics
was ranked second for midline of the
face and midline of the commissures.
The study protocol, however, did not
address the axial angulations of the
dental midline in its analysis. The re-
sults indicate that a symmetrical pat-
tern might exist in nature in arrang-
The Journal of Prosthetic Dentistry
ing the dental midline with respect
to the face and commissures. Thus,
it can be inferred that the incisive pa-
pilla, usually found in between the 2
maxillary central incisors, may be an
acceptable landmark for the deter-
mination of midlines, as reported by
authors in the past.6,7 Future studies
are needed to verify this. The philtrum
or tip of the vermillion border has
been assumed by several studies in
the past to represent the facial mid-
line.5,7,10-12,14,15 The present study
showed that the tip of the philtrum
ranked third in the hierarchy, super-
seded only by the midline of the com-
missures and the dental midline. This
reinforces the credibility of the tip of
the philtrum as a reliable landmark in
the determination of the midlines of
the face and mouth.
The nasion has been considered to
be a good location along the middle
fifth of the face, but its relation to the
facial and commissural midline has
not been studied previously. Based
on the current study, soft tissue na-
sion may not be an adequate clinical
landmark to determine either of the
midlines. Furthermore, its distant lo-
cation from the dental midline may
not result in easy determination and
analysis. The tip of the nose was the
most deviated landmark with regard
to the facial midline. However, it
ranked higher than the nasion with
regard to midline of the commissures.
Pearson correlations used to examine
the relationships of these 2 anatomic
landmarks showed that these 2 values
were similar; therefore, the reversal of
hierarchy could be due to a sampling
error and is not significant. Further-
more, both of these landmarks ranked
lowest in their proximity to the mid-
lines of the face and mouth; for most
clinical situations, the ranking of the
first 3 anatomic landmarks, namely,
the midline of the commissures, tip
of philtrum, and dental midline, ap-
pears to be more relevant.
The study used RFV and RCV (ra-
tios) as tools to examine the relation-
ship of the anatomic landmarks and
develop the hierarchy. Theoretically,
Volume 102 Issue 2
this could be a limitation in the study,
as ratios by nature cannot differenti-
ate whether the observed difference
is due to the numerator or denomi-
nator term. However, from a clinical
standpoint, the authors believe that
it is more important for a clinician to
know the hierarchy or the best choice
of anatomic landmarks that could be
used in the determination of midline
for a particular patient, rather than to
know mean linear deviations of ana-
tomic landmarks of a certain popula-
tion. Furthermore, the applied meth-
odology would not have permitted
sufficient accuracy to analyze linear
deviations, as the image dimensions
did not correspond to the exact di-
mensions of the subject’s face. The
authors did not report the direction
of deviation of each anatomic land-
mark with regard to the midlines, as
it has little clinical applicability. The
numerical values of the mean ratios
were different for the same landmark
depending upon the midline in ques-
tion; however, the hierarchy of land-
marks was not affected. Finally, the
study was done by a single observer
and the population chosen in this
study was based on convenience sam-
pling, with the sample distribution
being approximately normal. This
study provides baseline information
about the hierarchical relationships
of various facial anatomic landmarks
to the midlines of the face and mouth.
Similar studies on different samples
are needed to confirm the results.
CONCLUSIONS
Based on the limitations of this
study, the following conclusions were
drawn:
1. There was a significant differ-
ence between the mean ratios of the
chosen anatomic landmarks and the
midlines of the face and mouth.
2. The hierarchy of anatomic land-
marks closest to the midline of the
face is: (1) midline of the commis-
sures, (2) dental midline, (3) tip of
philtrum, (4) nasion, and (5) tip of
the nose.
Bidra et al
August 2009
3. The hierarchy of anatomic land-
marks closest to the midline of the
commissures is: (1) dental midline,
(2) tip of philtrum, (3) tip of the nose,
and (4) nasion.
REFERENCES
1. Rhodes G. The evolutionary psychol-
ogy of facial beauty. Annu Rev Psychol
2006;57:199-226.
2. Bashour M. History and current concepts
in the analysis of facial attractiveness. Plast
Reconstr Surg 2006;118:741-56.
3. Merriam-Webster’s medical dictionary.
Springfield: Merriam-Webster; 2006. Avail-
able at: http://www.nlm.nih.gov/medlin-
eplus/mplusdictionary.html. Accessed on
May 1, 2008.
4. Arnett GW, Bergman RT. Facial keys to
orthodontic diagnosis and treatment
planning--Part II. Am J Orthod Dentofacial
Orthop 1993;103:299-312.
5. Beyer JW, Lindauer SJ. Evaluation of
dental midline position. Semin Orthod
1998;4:146-52.
6. Zarb GA, Carlsson GE, Bolender CL, edi-
tors. Boucher’s prosthodontic treatment
for edentulous patients. 11th ed. St. Louis:
Mosby;1997. p. 309-10.
7. Latta GH Jr. The midline and its relation to
the anatomic landmarks in the edentulous
patient. J Prosthet Dent 1988;59:681-3.
8. Lombardi RE. The principles of visual
perception and their clinical applica-
tion to denture esthetics. J Prosthet Dent
1973;29:358-82.
Bidra et al
9. Tjan AH, Miller GD, The JG. Some es-
thetic factors in a smile. J Prosthet Dent
1984:51;24-8.
10.Arnett GW, McLaughlin RP. Facial and
dental planning for orthodontists and
oral surgeons. St. Louis: Elsevier; 2004. p.
52-70.
11.Kokich VO Jr, Kiyak HA, Shapiro PA. Com-
paring the perception of dentists and lay
people to altered dental esthetics. J Esthet
Dent 1999;11:311-24.
12.Johnston CD, Burden DJ, Stevenson MR.
The influence of dental to facial midline dis-
crepancies on dental attractiveness ratings.
Eur J Orthod 1999;21:517-22.
13.Rosenstiel SF, Rashid RG. Public prefer-
ences for anterior tooth variations: a
web-based study. J Esthet Restor Dent
2002;14:97-106.
14.Cardash HS, Ormanier Z, Laufer BZ.
Observable deviation of the facial and
anterior tooth midlines. J Prosthet Dent
2003;89:282-5.
15.Miller EL, Bodden WR Jr, Jamison HC.
A study of the relationship of the dental
midline to the facial median line. J Prosthet
Dent 1979;41:657-60.
16.Farkas LG. Anthropometry of the head and
face in medicine. 2nd Ed. New York: Raven
Press Limited; 1994. p. 20-5.
17.The glossary of prosthodontic terms. J
Prosthet Dent 2005;94:10-92.
18.American Association of Orthodontists
(AAO) Glossary 2008. Available at http://
www.braces.org/knowmore/glossary/
upload/2008-AAO-Glossary.pdf. Accessed
on May 1, 2008.
19.The American Academy of Facial Plastic
and Reconstructive Surgery. Glossary of
terms. Available at http://www.aafprs.org/
patient/procedures/glossary.html. Accessed
on May 1, 2008.
103
20.Owens EG, Goodacre CJ, Loh PL, Hanke
G, Okamura M, Jo KH, et al. A multicenter
interracial study of facial appearance. Part
1: A comparison of extraoral parameters.
Int J Prosthodont 2002;15:273-82.
21.Peng L, Cooke MS. Fifteen-year reproduc-
ibility of natural head posture: A longitudi-
nal study. Am J Orthod Dentofacial Orthop
1999;116:82-5.
22.Cooke MS. Five-year reproducibility of
natural head posture: a longitudinal
study. Am J Orthod Dentofacial Orthop
1990;97:489-94.
23.Cooke MS, Wei SH. The reproducibility of
natural head posture: a methodological
study. Am J Orthod Dentofacial Orthop
1988;93:280-8.
24.Lundström F, Lundström A. Natural head
position as a basis for cephalometric
analysis. Am J Orthod Dentofacial Orthop
1992;101:244-7.
Corresponding author:
Dr Avinash S. Bidra
University of Texas Health Science Center at
San Antonio
7703 Floyd Curl Drive, MSC 7912
San Antonio, Texas 78229-3900
Fax: 210-567-6376
E-mail: avinashbidra@yahoo.com
Acknowledgements
The authors thank Patricia Montgomery for
her support with the images and all partici-
pants who volunteered for the study.
Copyright © 2009 by the Editorial Council for
The Journal of Prosthetic Dentistry.